Preface

This book was created to explore the idea of dimensionality reduction as applied to the research question - how can one develop meaningful (culturally-relevant) groups in a given population. It consists of an initial assessment of all the dimensionality reduction algorithms for understanding culture relevant groupings dependent on personality and values Building this book has been an effort of love and perseverance towards the bigger goal of understanding the idea of Culture and if there is a way to quantify it. Much work needs to go into this before one could make conclusive arguments about which algorithm works best. This book should be viewed as more of a literature review of several high dimensionality algorithms, rather than a reference book of any sort. The ideas introduced here are still under development and should be considered with caution.

I would like to thank Gerard Saucier, Sanjay Srivastava, Sara Weston and Ryan Light for the insightful discussions on culture and its quantifiable properties. I am grateful to Daniel Anderson for his generous help with the coding and development of this book. I have immense gratitude for Daniel for his impeccable teaching skills and never ending patience. I would also like to thank Elizabeth McNeilly for her objective feedback on my crazy ideas.

Table of Contents

Chapter Name Description
Introduction 1 I begin by introducing the research question and its importance. I also introduce the Personality and Values dataset that will be used for demonstration purposes in the book. Finally, I discuss the organization of the book.
Cluster analysis 2 In this chapter I explore the algorithms that have been used to retain optimum number of clusters.
Network Analysis 3 I propose a network model of culture and an iterative process that can potentially be used for simulating real world conditions for group building.
Principal Component Analysis 4 Finally, I use PCA, a very common dimension reduction technique, and its application to the research question.